Advertisement

Pairwise Marginal Likelihood for the Bradley-Terry Model

Article

Abstract

The statistical inference for the Bradley-Terry model with logit link and random effects is often made cumbersome by the high-dimensional intractable integrals involved in the marginal likelihood. An inferential methodology based on the marginal pairwise likelihood approach is proposed. This method belongs to the broad class of composite likelihood and involves marginal pairs probabilities of the responses. These probabilities are derived from the approximation of the logistic distribution function by a normal scale mixture. The sample performance of this approach is evaluated by a simulation study. The main motivation of this work is the lizards data, on which the the proposed approach is illustrated.

Keywords

Bradley-Terry model Composite likelihood Fixed effects Lizards data Logit link Marginal pairwise likelihood Marginal maximum likelihood Probit link Variance component 

AMS Subject Classification

62F12 62H12 62J02 62J12 62J15 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Bellio, R., and C. Varin. 2005. A pairwise likelihood approach to generalized linear models with crossed effects. Stat. Model., 5, 217–227.MathSciNetCrossRefGoogle Scholar
  2. Bradley, R. A., and M. E. Terry. 1952. Rank analysis of incomplete block designs: The method of paired comparisons. Biometrika, 39, 324–345.MathSciNetMATHGoogle Scholar
  3. Besag, J. E. 1975. Statistical analysis of non lattice data. Statistician, 24(2), 179–195.CrossRefGoogle Scholar
  4. Cox, D. R., and N. Reid. 2004. A note on pseudolikelihood constructed from marginal densities. Biometrika, 91, 729–737.MathSciNetCrossRefGoogle Scholar
  5. Crouch, A., and D. Spiegelman. 1990. The evaluation of integrals of the form \(\int {f(t)} {e^{ - {t^2}}}\). Application to logistic-normal models. J. Am. Stat. Assoc., 85, 464–469.MathSciNetMATHGoogle Scholar
  6. Drum, M. L., and P. McCullagh. 1993. REML estimation with exact covariance in the logistic mixed models. Biometrics, 49, 677–689.MathSciNetCrossRefGoogle Scholar
  7. Feddag, M. L., and S. Bacci. 2009. Pairwise likelihood for the longitudinal mixed Rasch model. Comput. Stat. Data Anal., 53, 1027–1037.MathSciNetCrossRefGoogle Scholar
  8. Firth, D. 2004. Bradley-Terry scaling with explanatory variables and incomplete data. Manchester Seminar, November 2.Google Scholar
  9. Firth, D. 2005. Bradley-Terry model in R. J. Stat. Software, 12(1), 1–12.MathSciNetCrossRefGoogle Scholar
  10. Ford, L. R., Jr., 1957. Solution of a ranking problem from binary comparison. Am. Math. Monthly, 64, 28–33.MathSciNetCrossRefGoogle Scholar
  11. Kuk, Y. C., and D. N. Nott. 2000. A pairwise likelihood approach to analyzing correlated binary data. Stat. Probability Lett., 47, 329–335.CrossRefGoogle Scholar
  12. LeCessie, S., and J. C. van Houwelingen. 1994. Logistic regression for correlated binary data. App. Stat., 43, 95–108.CrossRefGoogle Scholar
  13. Lindsay, B. G. 1988. Composite likelihood methods. In Statistical inference from stochastic processes, Contemporary Mathematics 80, ed. N. U. Prablu, Providence, RI, American Mathematical Society. 221–239.CrossRefGoogle Scholar
  14. Monahan, J. F., and L. A. Stefanski. 1989. Normal scale mixture approximations to F*(z) and computation for the logistic-normal integral. In Handbook of the logistic distribution, ed. N. Balakrishnan, 529–540. New York, NY: Marcel Dekker.Google Scholar
  15. R Development Core Team. 2005. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. http://www.R-project.org.Google Scholar
  16. Renard, D., G. Molenberghs, and H. Geys. 2004. A pairwise likelihood approach to estimation in multilevel probit models. Comput. Stat. Data Anal., 44, 649–667.MathSciNetCrossRefGoogle Scholar
  17. Whiting, M. J., D.M. Stuart-Fox, D. O’Connor, D. Firth, N. C. Bennet, and S. P. Blomberg. 2006. Ultraviolet signals ultra-aggression in a lizard. Anim. Behav., 72, 353–363.CrossRefGoogle Scholar

Copyright information

© Grace Scientific Publishing 2013

Authors and Affiliations

  1. 1.EA 4275 “Biostatistique, Recherche Clinique et Mesures Subjectives en Santé,” Faculté de PharmacieUniversité de NantesNantesFrance

Personalised recommendations